The finding
A disaster grant from FEMA Individual Assistance is a few thousand dollars, not a repair budget. Across 25,944,188 owner and renter registrations covering 630 declared disasters from 2002 to 2026, 46.4% were approved and the mean approved grant is $3,331. 83.0% of approved grants fall in the $1 to $10,000 band and 0.80% reach the statutory maximum. That is what a capped, need-based programme pays.
Headline figures
In this report
Key findings
- The mean approved grant across both populations is $3,331, computed from $40,099,576,017 of approved assistance over 12,036,730 approvals.
- 46.4% of 25,944,188 registrations were approved, 46.8% for owners and 45.9% for renters.
- 83.0% of approved grants fall in the $1 to $10,000 band, 10.7% in $10,001 to $25,000 and 6.3% above $25,000.
- Only 96,563 grants reached the statutory maximum, 0.80% of all approvals, so the cap is binding on a small minority of cases.
- Of assistance dollars that are attributable to a component, 37.2% went to repair and replace, 21.5% to rental assistance and 41.3% to other needs.
- Mean grant varies 6.6 times across states, from $11,750 in Alaska to $1,769 in Puerto Rico, across the 39 states clearing the reporting floor.
- Approval rate varies 10.0 times across states, from 76.3% in Minnesota to 7.6% in Maine. This spread is the single widest variation in the ledger.
- Total approved assistance in the two files is $40,099,576,017, of which $25,451,448,848 went to owner-occupants and $14,648,127,169 to renters.
What this programme is and is not
Read this before the tables. Every figure below describes FEMA Individual Assistance, and that programme is not what a reader comparing it to a restoration estimate might assume it to be.
FEMA Individual Assistance is capped, need-based assistance for uninsured and underinsured necessary expenses and serious needs. It is not insurance. It does not reimburse the cost of restoring a property to its pre-loss condition, and it is not designed to. Statute sets a maximum award, assistance is intended to address immediate housing and serious needs, and private insurance and other sources are expected to carry losses they cover. Our SBA disaster loan report covers the federal loan channel that sits beside it.
So the fact that the mean approved grant is $3,331 while restoration work on a significantly damaged property runs into the tens of thousands is not a shortfall in programme performance. It is the arithmetic of a programme doing a narrower job than full indemnity. Reporting the figure is worthwhile because the gap is frequently misread in the other direction, as though an approved FEMA grant settles a restoration bill. It does not, and this report sets out the actual scale.
Nothing in this report is an assessment of FEMA, of any state agency or of any applicant. It is a statistical record of published assistance figures and the derived ratios that follow from them.
The national ledger
| Line | Owners | Renters | Both populations |
|---|---|---|---|
| Valid registrations | 15,121,818 | 10,822,370 | 25,944,188 |
| Approved for assistance | 7,070,433 | 4,966,297 | 12,036,730 |
| Approval rate | 46.8% | 45.9% | 46.4% |
| Total approved assistance | $25,451,448,848 | $14,648,127,169 | $40,099,576,017 |
| Mean approved grant | $3,600 | $2,950 | $3,331 |
| Repair and replace dollars | $13,985,589,157 | $13,034,614 | $13,998,623,770 |
| Rental assistance dollars | $3,914,101,524 | $4,170,211,917 | $8,084,313,441 |
| Other needs dollars | $6,273,384,269 | $9,289,336,807 | $15,562,721,076 |
| Grants at the statutory maximum | 80,032 | 16,531 | 96,563 |
Two figures in that table deserve a second look. The first is the mean grant of $3,331, which is the number most often needed and least often stated with its denominator. The second is that 13,907,458 registrations did not result in approved assistance. A registration is not a claim and non-approval covers several distinct situations, including insured losses, inspections finding no damage, and incomplete applications, none of which the published fields separate.
Where the money goes
| Component | Dollars | Share of attributable dollars | Owner share | Renter share |
|---|---|---|---|---|
| Repair and replace | $13,998,623,770 | 37.2% | 57.9% | 0.1% |
| Rental assistance | $8,084,313,441 | 21.5% | 16.2% | 31.0% |
| Other needs | $15,562,721,076 | 41.3% | 26.0% | 68.9% |
Repair and replace is the largest component at 37.2% of attributable dollars, and it is almost entirely an owner-occupant category: 57.9% of owner dollars against 0.1% of renter dollars. Rental assistance runs the opposite way. The split by tenure is set out in full in our owner renter damage gap report.
The component fields do not sum to the published total. Together they cover $37,645,658,287 of $40,099,576,017, or 93.9%. The residual is not attributable from the fields FEMA publishes in these two files, so every share above is expressed against the component total rather than against the headline figure. Shares computed against the headline figure instead would understate each component by roughly 6%.
Assistance mix by state
| State | Approved assistance | Repair share | Rental share | Other needs share |
|---|---|---|---|---|
| Louisiana | $8,823,457,670 | 24.9% | 39.8% | 35.3% |
| Florida | $6,179,527,326 | 34.5% | 20.4% | 45.1% |
| Texas | $5,178,499,841 | 41.8% | 16.4% | 41.9% |
| Puerto Rico | $2,627,947,644 | 33.9% | 7.6% | 58.4% |
| Mississippi | $1,848,102,559 | 33.5% | 25.1% | 41.4% |
| New York | $1,758,555,025 | 46.2% | 26.7% | 27.1% |
| Illinois | $1,415,413,546 | 51.9% | 22.6% | 25.6% |
| North Carolina | $1,046,346,833 | 43.3% | 10.7% | 46.0% |
| Michigan | $1,034,735,366 | 47.5% | 26.7% | 25.8% |
| New Jersey | $1,018,620,327 | 50.8% | 27.7% | 21.6% |
| California | $842,443,196 | 26.9% | 19.0% | 54.1% |
| Alabama | $704,554,304 | 38.0% | 18.6% | 43.4% |
| Georgia | $664,519,343 | 29.3% | 8.0% | 62.7% |
| Pennsylvania | $584,544,140 | 41.6% | 14.6% | 43.8% |
| South Carolina | $579,904,723 | 31.0% | 9.9% | 59.1% |
The mix shifts with tenure composition and with the kind of event. Louisiana runs 24.9% repair against 39.8% rental, while Florida runs 34.5% and 20.4%. For a restoration firm this is a read on whether the assistance flowing into a market is pointed at rebuilding or at rehousing, which is not the same demand signal.
Mean approved grant by state
| State | Approvals | Approved assistance | Mean approved grant | At statutory maximum |
|---|---|---|---|---|
| Alaska | 10,051 | $118,103,156 | $11,750 | 8.87% |
| North Dakota | 11,986 | $114,950,635 | $9,590 | 19.07% |
| Hawaii | 15,423 | $146,794,568 | $9,518 | 2.60% |
| Vermont | 10,404 | $70,654,395 | $6,791 | 3.83% |
| Kentucky | 71,522 | $475,672,061 | $6,651 | 2.66% |
| New York | 276,433 | $1,758,555,025 | $6,362 | 2.28% |
| Kansas | 10,012 | $61,093,294 | $6,102 | 3.40% |
| Washington | 15,140 | $89,515,330 | $5,913 | 2.67% |
| Oklahoma | 29,788 | $173,152,171 | $5,813 | 2.91% |
| California | 146,689 | $842,443,196 | $5,743 | 2.31% |
| Iowa | 54,797 | $312,084,769 | $5,695 | 3.44% |
| West Virginia | 50,209 | $276,917,048 | $5,515 | 1.61% |
| Minnesota | 15,321 | $83,580,246 | $5,455 | 1.66% |
| New Jersey | 194,563 | $1,018,620,327 | $5,235 | 1.31% |
| Maryland | 17,582 | $90,696,286 | $5,158 | 0.13% |
The spread is 6.6 times, from $11,750 in Alaska to $1,769 in Puerto Rico. Because assistance is need-based and capped, a higher state mean reflects the severity and type of the disasters that state has experienced and the uninsured share of those losses, accumulated across more than two decades of declarations. It is not a rate or a schedule that applies to a future event.
Grant band distribution and the statutory maximum
| Approved grant band | Grants | Share of graded grants | Owner share | Renter share |
|---|---|---|---|---|
| $1 to $10,000 | 9,991,977 | 83.01% | 74.28% | 95.44% |
| $10,001 to $25,000 | 1,282,008 | 10.65% | 15.29% | 4.05% |
| $25,001 to the statutory maximum | 762,745 | 6.34% | 10.43% | 0.51% |
| Measure | Owners | Renters | Both |
|---|---|---|---|
| Grants at the statutory maximum | 80,032 | 16,531 | 96,563 |
| As a share of approvals | 1.13% | 0.33% | 0.80% |
The maximum matters because it is the one point where the programme’s design cap, rather than assessed need, sets the award. 0.80% of approvals reached it, 96,563 grants in total. For the other 99.20% the award was below the cap, which means assessed need rather than the ceiling determined the amount.
Read alongside the band table, that is a consistent picture: 83.0% of graded grants sit in the $1 to $10,000 band, and the statutory maximum is reached rarely. The modal experience of this programme is a grant in the low thousands. That matches the modal damage event, which our restoration job size distribution report puts in the same dollar band.
The statutory maximum has been adjusted over the period these files cover, so totalMaxGrants counts grants that hit whichever cap applied at the time. The field cannot be converted into a single dollar figure across the whole period, and no such figure is quoted here.
Approval rate by state
| State | Registrations | Approvals | Approval rate | Mean approved grant |
|---|---|---|---|---|
| Minnesota | 20,079 | 15,321 | 76.3% | $5,455 |
| Massachusetts | 71,423 | 52,148 | 73.0% | $3,309 |
| Wisconsin | 174,066 | 126,072 | 72.4% | $3,890 |
| Illinois | 523,774 | 370,308 | 70.7% | $3,822 |
| West Virginia | 76,051 | 50,209 | 66.0% | $5,515 |
| Maryland | 27,603 | 17,582 | 63.7% | $5,158 |
| Ohio | 192,030 | 119,662 | 62.3% | $2,762 |
| Michigan | 503,712 | 313,684 | 62.3% | $3,299 |
| Rhode Island | 34,428 | 20,824 | 60.5% | $3,471 |
| Pennsylvania | 263,077 | 157,020 | 59.7% | $3,723 |
| Florida | 7,129,494 | 2,658,969 | 37.3% | $2,324 |
| Georgia | 793,069 | 280,996 | 35.4% | $2,365 |
| California | 551,883 | 146,689 | 26.6% | $5,743 |
| Oregon | 30,405 | 6,913 | 22.7% | $8,420 |
| Maine | 69,565 | 5,319 | 7.6% | $4,441 |
The spread here is the widest in the ledger: 76.3% in Minnesota against 7.6% in Maine, a factor of 10.0. Each state’s figure pools every declaration it has had since 2002, and declarations differ in hazard, in the share of affected households carrying insurance, and in how widely the declared area was drawn relative to the area actually damaged.
That last point is worth stating plainly, because it is the most likely driver of the low end of this table. A declaration covering a large population generates registrations from households that turn out to be undamaged, and those registrations sit in the denominator. Approval rate is therefore a composite of eligibility and of declaration geometry, and the published fields do not allow the two to be separated. No inference about any state’s administration follows from this table.
Method
How we counted. We retrieved 160,390 rows from FEMA OpenFEMA HousingAssistanceOwners v2 and 147,970 rows from HousingAssistanceRenters v2 on 3 October 2026. Rows are FEMA aggregates by declared disaster, state, county, city and postcode, so every figure is a sum of fields rather than a count of rows. Mean approved grant is summed totalApprovedIhpAmount divided by summed approvedForFemaAssistance, computed at national and state level; it is not an average of per-row means, which would weight a three-approval postcode equally with a county of thousands. Approval rate is summed approvedForFemaAssistance over summed validRegistrations. Component shares use the sum of repairReplaceAmount, rentalAmount and otherNeedsAmount as the base, because those three fields sum to 93.9% of totalApprovedIhpAmount rather than to 100%, and the residual is not attributable from the published fields. Grant band shares use the sum of the three band counts as the base. The statutory-maximum share uses totalMaxGrants over approvedForFemaAssistance. State tables require at least 10,000 approvals for grant-level metrics and 20,000 registrations for approval rate; territory codes are excluded from state tables and retained in national totals. No inflation adjustment is applied, and the statutory maximum changed during the covered period.
Reproduce this
Reproduce this
- Endpoint
https://www.fema.gov/api/open/v2/HousingAssistanceOwners and https://www.fema.gov/api/open/v2/HousingAssistanceRenters- Query
$top=5000 paged with $skip- Rows retrieved
- 160,390 owner rows and 147,970 renter rows
- Retrieved
- 3 October 2026
Both endpoints page at 5,000 rows via $top and $skip. No API key is required. The two files must be pulled separately because their damage fields differ, although the assistance fields used throughout this report are identical in both.
What this does not show
What this does not show
- FEMA assistance is not insurance and not a restoration-cost reimbursement. It is capped, need-based assistance for uninsured and underinsured necessary expenses. The mean grant should not be read as the cost of repairing the damage.
- Approved applicants only. Every dollar figure here describes the 12,036,730 approvals. The 13,907,458 registrations that were not approved contribute to approval rate and to nothing else, and households that never registered are absent entirely. This dataset cannot say anything about them.
- Non-approval is not explained by the data. The published fields do not separate insured losses, inspections finding no damage, withdrawn applications and incomplete applications. The approval rate is a composite.
- Declared disasters only. Undeclared events, which are the large majority of water and fire losses, never enter these files.
- No per-applicant record. Rows are pre-aggregated, so no distribution within a grant band and no median grant can be recovered. Only means are available.
- The statutory maximum changed over the period. totalMaxGrants counts grants that reached whichever cap was in force at the time, so the count spans several different dollar ceilings.
- Component dollars do not sum to the total. The three components cover 93.9% of approved assistance. The residual is not attributable from these files.
- No inflation adjustment. All dollar figures span 2002 to 2026 in nominal terms, so the national mean is weighted toward the years with the largest disasters rather than being a constant-dollar average.
Most quotable figures
- The mean approved FEMA Individual Assistance housing grant is $3,331 across 12,036,730 approvals, 2002 to 2026 (FEMA OpenFEMA Housing Assistance v2, retrieved 3 October 2026).
- 83.0% of approved FEMA housing grants fall in the $1 to $10,000 band, and only 0.80% reach the statutory maximum (FEMA OpenFEMA, retrieved 3 October 2026).
- 46.4% of 25,944,188 FEMA housing registrations were approved for assistance, with state rates ranging from 76.3% to 7.6% (FEMA OpenFEMA, retrieved 3 October 2026).
- Of attributable FEMA housing assistance dollars, 37.2% went to repair and replace, 21.5% to rental assistance and 41.3% to other needs (FEMA OpenFEMA, 2002 to 2026).
Questions this report answers
How much is the average FEMA disaster grant?
Across owner and renter housing assistance records covering 2002 to 2026, the mean approved grant is $3,331, computed from $40,099,576,017 over 12,036,730 approvals. The owner mean is $3,600 and the renter mean is $2,950.
Does FEMA pay for restoration work?
Not as a cost reimbursement. FEMA Individual Assistance provides capped, need-based assistance toward uninsured and underinsured necessary expenses and serious needs. It is not insurance and is not designed to restore a property to its pre-loss condition.
How often does a grant reach the maximum?
Rarely. 96,563 grants reached the statutory maximum, which is 0.80% of approvals. For the rest, assessed need rather than the cap set the amount.
What share of applications is approved?
46.4% of 25,944,188 valid registrations, 46.8% for owners and 45.9% for renters. State rates vary widely, partly because a declaration drawn over a large population generates registrations from undamaged households.
What does the assistance actually pay for?
Of attributable dollars, 37.2% went to repair and replace, 21.5% to rental assistance and 41.3% to other needs. The mix differs sharply between owner-occupants and renters.
Why does the mean grant vary so much by state?
Assistance is need-based, so a state’s mean reflects the severity and type of disasters it has experienced since 2002 and the uninsured share of those losses. The range runs from $11,750 in Alaska to $1,769 in Puerto Rico.
Can I reuse these figures?
Yes. The tables and derived figures are released under CC BY 4.0 with attribution to Remediation Insider. The underlying FEMA data is public and reproducible from the endpoints listed above.
How to cite
Remediation Insider. The Disaster Grant Ledger: What FEMA Assistance Actually Pays. https://remediationinsider.com/disaster-grant-ledger/ (retrieved 2 October 2026).
Data and tables in this report are released under CC BY 4.0. Attribution: Remediation Insider.
Sources
- Federal Emergency Management Agency. FEMA OpenFEMA Housing Assistance, v2. HousingAssistanceOwners, 160,390 aggregated rows, retrieved 3 October 2026.
- Federal Emergency Management Agency. HousingAssistanceRenters v2, 147,970 aggregated rows, retrieved 3 October 2026.
- FEMA Individual Assistance Program overview, for the statutory basis and scope of the programme.
- FEMA OpenFEMA Terms and Conditions.
- Field definitions are taken from the FEMA OpenFEMA data dictionary for both Housing Assistance datasets.
- No survey, interview or physical inspection was conducted for this report. Every figure is computed at build time from the retrieved FEMA files.
This product uses the Federal Emergency Management Agency’s OpenFEMA API, but is not endorsed by FEMA. The Federal Government or FEMA cannot vouch for the data or analyses derived from these data after the data have been retrieved from the Agency’s website(s).